In recent years, many different techniques for the fast production of three-dimensional models have been developed for industrial use, which are sometimes referred to as rapid prototyping and manufacturing ("RP&M") techniques. In general, rapid prototyping and manufacturing techniques build three-dimensional objects, layer-by-layer, from a working medium utilizing a sliced data set representing cross-sections of the object to be formed. Typically an object representation is initially provided by a Computer Aided Design (CAD) system.
Stereolithography, the presently dominant RP&M technique, may be defined as a technique for the automated fabrication of three-dimensional objects from a fluid-like material utilizing selective exposure of layers of the material at a working surface to solidify and adhere successive layers of the object (i.e. laminae). In stereolithography, data representing the three-dimensional object is input as, or converted into, two-dimensional, layer data representing cross-sections of the object. Layers of material are successively formed and selectively transformed (i.e., cured) into successive laminae according to the two-dimensional layer data. During transformation, the successive laminae are bonded to previously formed laminae to allow integral formation of the three-dimensional object.
Though, stereolithography has shown itself to be an effective technique for forming three-dimensional objects, various improvements have been desired for some time. Many improvements have been made to object accuracy over the years; however, there still remains a need for improving accuracy further. Various aspects of the stereolithographic building process can impact the accuracy of objects formed. For instance, one property of the resins currently used in this process is that the resins are prone to shrinkage during solidification. FIG. 1 illustrates this problem.
FIG. 1 depicts a series of layers, L(i), L(2) . . . L(I), solidified on top of each other. As seen in FIG. 1, the lower layers, e.g., L(1), L(2), particularly, the unsupported portions, are bent upward towards the subsequently formed layers. The bending occurs because as each of the subsequent layers solidify, the newly formed laminae shrink. As each laminae shrinks, it pulls the laminae below upward and causes distortion to the object.
Various techniques involving the utilization of certain exposure strategies have been utilized to overcome this shrinkage/distortion problem. Examples of these techniques may be found in U.S. Pat. Nos. 5,104,592 and 5,256,340. One such technique involves the use of multiple exposures in the solidification of individual laminae wherein the first exposure might not cause direct adhesion to previous lamina. In addition to the vertical application of this technique, discussed above, the technique could be applied in a horizontal application by forming spatially separated regions on a lamina followed by the complete or partial solidification of the intermediate regions during a subsequent exposure of that lamina or during exposure of a subsequent lamina.
Due to advances in solid state laser development, the stereolithographic art has recently started to turn away from the inefficient gas lasers commercially used in the past and has begun to turn to frequency-multiplied, solid-state lasers. Frequency tripling of 1049 nm-1064 nm Nd/NYAG, Nd/YVO.sub.4, and Nd/YLF lasers produces wavelengths of 355 nm (YAG and YVO.sub.4), 351 nm (YLF) and 349 nm (YLF), which are all suitable for use in stereolithography with current resin formulations. Frequency quadrupling of 1342 nm Nd/YVO.sub.4 lasers produces a wavelength (335 nm) which is suitable for stereolithography as well. More detail about solid-state lasers can be found in U.S. application, Ser. No. 08/792,347. As applied to stereolithography up to this point in time, these lasers operate in a constant-repetition, pulsed mode.
As the stereolithographic art has turned to these new lasers and benefited from their increased efficiency and longer life, building styles and exposure techniques have not yet been optimized. A need for optimizing the stereolithography process for use with pulsed laser sources exists. This particularly applies to optimizations that will allow distortion reduction techniques, like those described above, to be utilized in combination with these pulsed lasers.
Another problem encountered during the building process relates to the ability to obtain a uniform exposure of the layer of solidifiable material. Currently, in some systems, the layers are solidified by scanning a light beam in pre-determined scanning lines across a surface of the liquid (e.g. the upper surface which is located at a working or target surface). Conventionally, a scanner is used consisting of two rotating mirrors deflecting the incident light beam in x- and y-directions to trace the desired object contour and to fill interior portions of the respective laminae.
Due to the velocity changes in the mirrors, some portions of the scan path may be subjected to a greater amount of radiation than other portions. This variation in applied radiation, i.e. exposure, may result in a variation in solidification depth induced in the medium. The simplified case of a single linear scanning line exemplifies this problem. To solidify the line, the scanning mirrors position the laser beam spot at the beginning of the scanning line, where the velocity of the beam spot may be zero. Thereafter, the scanning mirrors accelerate until they, and thus the beam spot, reach the desired velocity. When the end of the scanning line is reached, a deceleration of the beam spot occurs until its velocity may again be zero.
Depending upon the inertia of the scanning mirrors, the acceleration and deceleration phases might be a significant portion of the total distance moved through the scanned line. As stated above, during these phases, the laser-beam spot velocity is continuously changing. If, as in a conventional SLA, a continuously operating laser system is used, which emits a light beam of constant intensity, the exposure (defined as the product of the intensity and the effective residence time of each portion of the beam on a given unit area of the material) at each position along the scan line will change inversely in proportion to the changing beam spot velocity. Thus, in the simplified case, a nonuniform cure depth will occur in all portions of the scanning path where the velocity of the mirrors is accelerating or decelerating, e.g., at the beginning and ends of the scan path.
Unlike the simplified case of a linear line, in practice, the beam spot velocity does not necessarily reach zero at the ends of the scanning lines, rather, there may be a sudden turn in the beam spot path. Nonetheless, as in the simplified example, the adverse effects of nonuniform cure depth still may occur whenever the velocity of the mirrors change.
FIGS. 2a-2d depict an example for the simplified case. These Figures depict an exemplary relationship, for a line being scanned, between the intensity, I, of the beam (FIG. 2a), the velocity, v, of the beam (FIG. 2b), the exposure level, E, (FIG. 2c), and the resulting cure profile, CP, (FIG. 2d) at each point along a line being scanned in the x-direction. The exposure which is related to the ratio of the intensity and the velocity varies as the beam spot accelerates and decelerates. In the acceleration phase, AP, the velocity of the beam spot increases as a function of time and results in the exposure of the fluid medium at a given position decreasing as a function of time. In the deceleration phase, DP, as the velocity of the beam spot decreases, the exposure of the fluid medium at a given position increases. The uneven exposure of the fluid medium depicted in FIG. 2c may result in the nonuniform cure profile shown in FIG. 2d.
U.S. Pat. No. 5,014,207 to Lawton proposes an approach to overcome this problem. This patent is hereby incorporated by reference herein as if set forth in full. In the approach proposed by Lawton, the intensity of the laser beam is modulated from a substantially zero-intensity-level to a maximum intensity level through a modulation means, such as, an acousto-optical modulator. Further, the beam spot velocity is measured and a control computer adjusts the intensity of the light beam in proportion to the measured velocity such that a constant exposure level results. This approach presents various problems. For instance, to be of optimal use it requires a real-time determination of the velocity of the beam spot such that an instantaneous modulation of the intensity can be made.
Regardless of the above technique for obtaining uniform exposure as taught in the '207 patent, a need for optimizing the uniformity of exposure in stereolithography when using pulsed laser sources still exists. Uneven exposure still exists in pulsed laser stereolithography systems. This is due, in part, to the constant pulse repetition rate of the pulsed laser as currently utilized in stereolithography systems.
As shown in FIGS. 3a-3c, the problem of a greater exposure at the beginning and end of the scan lines is also present in the case of constant repetition rate pulsed lasers. FIG. 3a depicts the laser pulses as a function of time. As shown, the pulses are constant and evenly spaced. FIG. 3b depicts the resulting pulses of FIG. 3a as a function of position. As seen in FIG. 3b, an increased density of pulses occurs at the positions scanned during the acceleration and deceleration phases. The cure profile resulting from the constant laser pulse is shown in FIG. 3c. A comparison of FIG. 3c to FIG. 2d reveals the similarity between the cure profiles in the case utilizing a continuous wave laser (i.e., cw laser) and the case utilizing a constant-repetition-rate pulsed laser. As can be seen from the nonuniform cure profiles, there is a need in the industry to eliminate the dependency of the exposure of the liquid medium from the scanning speed.
The problems described herein above are addressed singly and/or in combination by the invention and are illustrated in the embodiments to be discussed herein after.